On Thu, Sep 24, 2009 at 12:04:54PM +0200, Henrik Sarvell wrote:
> Yes you should revert back, and I suppose the best solution is to
> implement the change you were talking about above. How much overhead
> would it introduce?
Perhaps not much overhead, but it needs another quirky rule, like
While reading characters of an atom, as long as the result looks like
a number, allow the dot, otherwise not
The problem is that Lisp, as opposed to other languages, allows symbols
to begin with a digit or a sign character (e.g. '1+" or '-123symbol').
So, for a trivial solution , '-12.3symbol' would allow the dot despite
it is not a number, while 'a.b' would split at the dot. This is not
Better would be first to read the whole atom, analyse it, and then
decide whether it is a number or not. This would break the current
simple single-character-look-ahead algorithm, though.
So for now I would tend to stay with Tomas' proposal, in handling the
dot as a meta-character only when not part of an atom (i.e. surrounded
by white space or other meta-characters like '(' and ')'). The advantage
is that then we can use '.' as part of symbol names, which is quite nice